Global Reprogramming of Host Kinase Signaling in Response to Fungal Infection

Abstract

Cryptococcus neoformans (Cn) is a deadly fungal pathogen whose intracellular lifestyle is important for virulence. Host mechanisms controlling fungal phagocytosis and replication remain obscure. Here, we perform a global phosphoproteomic analysis of the host response to Cryptococcus infection. Our analysis reveals numerous and diverse host proteins that are differentially phosphorylated following fungal ingestion by macrophages, thereby indicating global reprogramming of host kinase signaling. Notably, phagocytosis of the pathogen activates the host autophagy initiation complex (AIC) and the upstream regulatory components LKB1 and AMPKα, which regulate autophagy induction through their kinase activities. Deletion of Prkaa1, the gene encoding AMPKα1, in monocytes results in resistance to fungal colonization of mice. Finally, the recruitment of AIC components to nascent Cryptococcus-containing vacuoles (CnCVs) regulates the intracellular trafficking and replication of the pathogen. These findings demonstrate that host AIC regulatory networks confer susceptibility to infection and establish a proteomic resource for elucidating host mechanisms that regulate fungal intracellular parasitism.

Keywords: AMPKα; Cryptococcus neoformans; autophagy initiation complex; fungal pathogen; host factors; host-pathogen interaction; kinase signaling; phagocytosis; phosphoproteomics; proteomics.

Figure 1. Cn infection induces global reprograming of the host phosphoproteome

(a) RAW264.7 macrophages were infected with antibody-opsonized Cn-dsRed. At 3 hrs (top panels) or 24 hrs (bottom panels) post-infection, the intracellular and extracellular Cn-dsRed populations were determined by staining unpermeabilized infected host cells. The stained cells were then processed for differential interference contrast (DIC) microscopy (left panels) or fluorescence imaging (right panels) (60X). The extracellular population of Cn-dsRed cells was stained green (right panels). (b) Work flow for phosphoproteomic analysis of mock or Cn infected RAW264.7 cells (n=60). (c) Total numbers of phosphoproteins and phospho-peptides identified by LC-MS/MS analysis. The numbers of phosphoproteins and phosphopeptides that displayed 1.5 fold or greater enhancement (up) or reduction (down) in the levels of phosphorylation are also shown. Log₂ curve of the peptide (d), or phosphopeptide (e), abundance ratios for Cn infected cells vs. mock-infected controls. Abbreviations, Cn: Cryptococcus neoformans, h.p.i.: Hours post infection, MOI: Multiplicity of infection, DIC: Differential interference contrast, LC-MS/MS: Liquid chromatography-tandem mass spectrometry. See also, Figures S1–S2, S7 and Tables S1–S2.

Figure 2. Cn induces activation of AMPK network

KEGG maps were generated to represent mTOR and AMPK network components that display responsiveness to Cn infection as revealed by phosphoproteomic or motif-X analyses. Abbreviations: Cn: Cryptococcus neoformans; KEGG: Kyoto Encyclopedia of Genes and Genomes. See also, Figures S1–S2 and Tables S1–S2, S7.

Figure 3. Cn infection induces activation of AIC signaling in host cells

(a–c) Western blot (bottom) and image J-based quantification (top) of host LKB1, AMPKα and ULK1 phosphorylation during a time course of Cn infection of RAW 264.7 cells. (d–j) Effects of compounds that modulate AMPKα activity on Cn-induced phosphorylation of host AMPKα and ULK1. Western blot (bottom) and quantification (top) was performed on cell lysates to detect and quantify the phosphorylation of AMPKα (Thr172) and/or ULK1 (Ser555). (j) Activation of host ribosomal protein S6 (Ser235/236) by Cn infection. RAW 264.7 cells were infected with Cn. At 3 h.p.i., the infected cells were processed for flow cytometry to assess rpS6 (Ser235/236) activation. A representative overlay histogram and the average of the Mean Flourescence Intensities (MFI) is shown (n=4). Black and blue histograms represent mock-infected and Cn-infected groups, respectively. (k) Cn-induced phosphorylation of AMPKα is dependent upon LKB1 and AIC components. RAW264.7 macrophages depleted of LKB1, ATG13, FIP200 or ULK1 were infected with Cn. At the indicated time points post-infection, lysates from the infected cells were analyzed by Western blot using p-AMPK (Thr172) or pan-AMPK antibodies. Results from one representative experiment (n=5) are shown. (l, m) CD11b⁺ cells from splenocytes of WT mice were mock-infected or infected ex vivo with Cn for 30 min. The levels of phosphorylation of AMPKα or ULK1 were then determined by flow cytometry using antibodies directed against pAMPKα1 (Thr172) or pULK1 (Ser555), respectively. Black, blue and green histograms represent mock-infected, Cn-infected and AMPK activator berberine (BBR)-treated groups (n=4 per group), respectively. (n, o) In vivo activation of AMPKα or ULK1 in CD11b⁺ cells from the spleens of mock-infected or infected WT mice at 6 h.p.i. was determined by phospoflow cytometry using anti-phospho antibodies against pAMPKα1 (Thr172) or pULK1-(Ser555), respectively. Representative overlay histograms and average MFIs are displayed (n=5 per group). Black and blue histograms represent mock-infected and Cn-infected groups, respectively. Images were from one representative experiment (n=3). Abbreviations: AICAR: 5-Aminoimidazole-4-carboxamide ribonucleotide; BBR: berberine; CC: Compound C; Cn: Cryptococcus neoformans; MFI: mean fluorescence intensity; MOI: multiplicity of infection; m.p.i: minutes post-infection; PBS: phosphate buffered saline; WB: Western blot; WT: wild-type. See also, Figures S3–S4, S7.

Figure 4. Depletion or inhibition of host AIC components disrupted the intracellular lifestyle of Cn

(a) ATG13-depleted or LKB1-depleted host cells show resistance to Cn infection. Scrambled control, ATG13-depleted, or LKB1-depleted RAW264.7 cells were infected with Cn-dsRed. At 2 h.p.i., the extracellular Cn population was visualized by immunofluorescence staining of unpermeabilized cells. White arrows indicate unstained intracellular Cn-dsRed cells. Yellow arrows indicate extracellular Cn-dsRed cells. (b–f) shRNA-mediated depletion or pharmacological inactivation of host AMPKα, AMPKβ, ATG13, FIP200, LKB1, ULK1 or Atg9a reduced the phagocytosis of antibody opsonized Cn by RAW264.7 or B6J2 macrophages, as indicated. Phagocytosis of Cn was measured using immunofluorescence microscopy (IFM) analysis to enumerate the number of internalized Cn per host cell at 2 or 24 h.p.i. (b, c), by fluconazole protection/colony formation unit analysis (CFU) to determine relative internalization (d–f) (Qin et al., 2011), or by flow cytometry (g, h) (Nicola et al., 2011). Representative dot plots (g) and histograms (h) of the flow cytometry findings for the indicated times post-infection are shown. Data represent the mean and standard deviation (SD) from six independent experiments (b–f). *, Indicates significance at P<0.05; **, Indicates significance at P<0.01, ***, Indicates significance at P<0.001. (i) Pretreatment of host cells with AMPKα inhibitor Compound C (CC, 20μM) or activators AICAR (20μM), BBR (20μM), (j) salicylate (10mM), A769662 (100μm) or (k) rapamycin (10μm) significantly decreased or increased, respectivey, Cn phagocytosis. *, indicates significance at P<0.05; **, indicates significance at P<0.01. Abbreviations: AICAR: 5-Aminoimidazole-4-carboxamide ribonucleotide; BBR: berberine; CC: Compound C; CMFDA: 5-chloromethylfluorescein diacetate; Cn: Cryptococcus neoformans; Ctrl: control; DDAO-SE: 9- H-(1,3-dichloro-9,9-dimethylacridin-2-one-7-yl)-succinimidyl ester; MOI: multiplicity of infection; h.p.i: hours post-infection; RAPA: rapamycin; SD: standard deviation; WT: wild-type. See also, Figures S3–S4, S7

Figure 5. AIC and AIC regulatory components are coordinately recruited to forming phagosomes and nascent CnCVs

(a) Immunofluoscence microscopy was used to analyze the recruitment and colocalization of AIC and AIC regulatory proteins (host AMPKα, ULK1, ATG13, FIP200 and ATG9a) on forming phagosomes and nascent CnCVs in RAW264.7 macrophages at 2 h.p.i. (b) Representative live cell fluorescence microscopy image of Atg9a-GFP recruitment to forming phagosomes containing Cn. The yellow arrows indicate regions of notable Atg9-GFP recruitment. (c) Representative immunofluorescence micrograph demonstrating the recruitment of host AMPKα to forming phagosomes that contain Cn. The yellow arrows indicate regions of notable AMPKα recruitment. (d) Representative transmission electron micrograph of a Cn-infected RAW264.7 macrophage showing a CnCV circumscribed by a single host membrane. (e) Immunofluorescence micrographs depicted the diminished ATG9a recruitment to nascent CnCVs in host cells depleted of ULK1 compared to non-depleted controls. (f) Quantification of the amount of ATG9a recruitment to forming pathogen-containing phagosomes or nascent CnCVs in RAW264.7 macrophages depleted of the indicated AIC proteins. Data represent the means ± standard deviations from three independent experiments. ** indicates significance at P<0.01. Scale bar: 5 μM. Abbreviations: Cn: Cryptococcus neoformans; CnCV: Cryptococcus neoformans-containing vacuole; Ctrl: control. See also, Figure S7

Figure 6. Host AMPKα activity confers susceptibility to Cn infection of primary cells and mice

(a) Western blot analysis of AMPKα (Prkaa1) protein expression in control (Ctrl) and Prkaa1 CKO mice harboring deletions in the gene encoding AMPKα. (b) Murine BMDMs pre-treated with CC (20μM) displayed resistance to pathogen internalization. In addition, AMPKα1^−/− (CKO) BMDMs displayed striking resistance to phagocytosis (2 h.p.i.) or intracellular replication (24 h.p.i) of the pathogen compared to AMPKα1^+/+ controls. (c) Treatment of mice with the AMPKα inhibitor compound C (CC) reduced Cn tissue burden in infected animals. Fungal burden in the indicated tissues was analyzed at 6 h.p.i. by plating homogenized tissue dilutions on solid medium (YPD-agar). Data represent CFU/ml, mean ± SD from 3 replicate experiments each with 5 mice/group. *, indicates significance at P<0.05. (d, e) Immunological characterization of Prkaa1 CKO mice. Splenocytes from control (Ctrl) and CKO mice were stained with fluorescently labeled antibodies against CD19, CD4, CD8 and CD11b to assess the percentage of B cells, T cell subsets and CD11b⁺ myeloid cells, respectively, in the spleen of CKO and the control mice by flow cytometry. Data represents mean ±SD of percentage of indicated cells in total splenocytes from 5 mice. (f, g) Cn burden in the tissues of Cn-infected CKO or littermate control mice at 6 h.p.i or 14 d.p.i. Mice from each group were infected with 10⁵ Cn via intra-pulmonary instillation and the fungal burden in the indicated tissues was determined at 6 h.p.i. (f) or 14 d.p.i. (g). Data represent the mean percentage of tissue fungal burden in the control. Data were from three independent experiments with 5 mice per group in each experiment. *, Indicates significance at P<0.05. (h) Cn burden in CD11b⁺ myeloid cells isolated by magnetic separation from spleocytes of Cn-infected (i.p.) CKO or control mice at 14 d.p.i. Data represent the mean ± SD CFU/ml from 5 mice. *, Indicates significance at P<0.05. (i) Histopathology of representative hematoxylin and eosin (H&E) stained sections of tissues (liver, panels 1, 2; lung panels 3, 4) from Cn-infected control (panels 1, 3) or CKO (panels 2, 4) mice at 14 d.p.i. In the liver (1, 2), inflammatory foci were randomly distributed (black arrowheads) but were more numerous and larger in control animals. Panel 1 inset shows detail of an inflammatory focus. In the images of lung tissue (3, 4), the yellow arrowheads point to fungal organisms within the alveolar spaces. While most alveoli in control mice contained fungi, the alveoli of CKO mice were often clear of pathogen (green arrowheads). The red arrowheads demonstrate that the interstitium of the lung was infiltrated with inflammatory cells and thicker in control animals. Panel 3 inset: a low magnification (4X) image of inflammed pulmonary tissue. Panel 4 Inset: a high magnification of pulmonary tissue with alveolar spaces clear of fungi. Abbreviations: CFU: colony forming units; CKO: conditional knockout; Ctrl: Control; Cn: Cryptococcus neoformans; h.p.i.: hours post-infection; d.p.i.: days post-infection; SD: standard deviation. See also, Figures S5–S7.